Automatic choice of the threshold of a grain filter via Galton-Watson trees. Application to granite cracks detection

Romain Abraham (1) , Maïtine Bergounioux (1) , Pierre Debs (1)

1 MAPMO - Mathématiques - Analyse, Probabilités, Modélisation - Orléans

The goal of this paper is the presentation of a post-processing method allowing to remove impulse noise in binary images, while preserving thin structures. We use a grain filter as in [5]. We propose a method to automatically determine the required threshold using Galton-Watson processes. We present numerical results and a complete analysis on a synthetic image. We end the numerical section considering a specific application to granite samples crack detection: here we deal with X-tomography images that have been binarized via preprocessing techniques and we want to remove residual impulse noise while keeping cracks and micro-cracks structure.

Mots clés: Image processing; Galton-Watson; grain filter; cracks

Domaines: Probabilités [math.PR]; Traitement des images [eess.IV]

Romain Abraham, Maïtine Bergounioux, Pierre Debs. Automatic choice of the threshold of a grain filter via Galton-Watson trees. Application to granite cracks detection. Journal of Mathematical Imaging and Vision, 2018, 6 (1), pp.50-69. ⟨hal-01337551v2⟩